Hydrogen induced structure and property changes in Eu3Si4

Article abstract of DOI:10.1016/j.jssc.2019.05.033

Hydrides Eu₃Si₄H_2+x were obtained by exposing the Zintl phase Eu₃Si₄ to a hydrogen atmosphere at a pressure of 30 bar and temperatures from 25 to 300 °C. Structural analysis using powder X-ray diffraction (PXRD)data suggested that hydrogenations in a temperature range 25–200 °C afford a uniform hydride phase with an orthorhombic structure (Immm, a ≈ 4.40 ?, b ≈ 3.97 ?, c ≈ 19.8 ?), whereas at 300 °C mixtures of two orthorhombic phases with c ≈ 19.86 and ≈ 19.58 ? were obtained. The assignment of a composition Eu₃Si₄H_2+x is based on first principles DFT calculations, which indicated a distinct crystallographic site for H in the Eu₃Si₄ structure. In this position, H atoms are coordinated in a tetrahedral fashion by Eu atoms. The resulting hydride Eu₃Si₄H₂ is stable by ?0.46 eV/H atom with respect to Eu₃Si₄ and gaseous H₂. Deviations between the lattice parameters of the DFT optimized Eu₃Si₄H₂ structure and the ones extracted from PXRD patterns pointed to the presence of additional H in interstitials also involving Si atoms. Subsequent DFT modeling of compositions Eu₃Si₄H₃ and Eu₃Si₄H₄ showed considerably better agreement to the experimental unit cell volumes. It was then concluded that the hydrides of Eu₃Si₄ have a composition Eu₃Si₄H_2+x (x < 2)and are disordered with respect to H in Si₂Eu₃ interstitials. Eu₃Si₄ is a ferromagnet with a T_C at about 120 K. Ferromagnetism is effectively quenched in Eu₃Si₄H_2+x. The effective magnetic moment for both materials is 7.5 μ_B which is typical for compounds containing Eu²⁺ 4f⁷ ions.

Full text of DOI:10.1016/j.jssc.2019.05.033

Accepted Manuscript
Hydrogen induced structure and property changes in Eu Si
3
4
Gustav Ek, Reji Nedumkandathil, Robert Johansson, Jorge Montero, Claudia Zlotea,
Mikael S. Andersson, Per Nordblad, Chiu Tang, Martin Sahlberg, Ulrich Häussermann
PII:
S0022-4596(19)30262-2
https://doi.org/10.1016/j.jssc.2019.05.033
YJSSC 20776
DOI:
Reference:
To appear in: Journal of Solid State Chemistry
Received Date: 20 March 2019
Revised Date: 19 May 2019
Accepted Date: 19 May 2019
Please cite this article as: G. Ek, R. Nedumkandathil, R. Johansson, J. Montero, C. Zlotea, M.S.
Andersson, P. Nordblad, C. Tang, M. Sahlberg, U. Häussermann, Hydrogen induced structure and
property changes in Eu Si , Journal of Solid State Chemistry (2019), doi: https://doi.org/10.1016/
3
4
j.jssc.2019.05.033.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
2
1
0
Eu Si
Eu3Si4H2+x
3
4
c
b
a
Eu Si
3
4
Eu Si _H _200C
3
4
2
0
50
100
150
200
T ꢀK)

ACCEPTED MANUSCRIPT
Hydrogen Induced Structure and Property Changes in Eu Si
3
4
a
b
c
d
d
Gustav Ek , Reji Nedumkandathil , Robert Johansson , Jorge Montero , Claudia Zlotea ,
e,**
e
f
a
b,*
Mikael S. Andersson , Per Nordblad , Chiu Tang , Martin Sahlberg , Ulrich Häussermann
a
Department of Chemistry - Ångström laboratory, Uppsala University, Sweden
b
Department of Materials and Environmental Chemistry, Stockholm University, Sweden
c
Department of Physics and Astronomy, Uppsala University, Sweden
d
Université Paris Est, Institut de Chimie et des Matériaux Paris-Est, CNRS, France
e
Department of Engineering Sciences, Uppsala University, Sweden
f
Diamond Light Source, Harwell Science and Innovation Campus, UK
*
Ulrich.Haussermann@mmk.su.se
**
Present address: Department of Chemistry and Chemical Engineering, Chalmers University
of Technology, Sweden
1

ACCEPTED MANUSCRIPT
Abstract
Hydrides Eu Si H
were obtained by exposing the Zintl phase Eu Si to a hydrogen
3
4
2+x
3
4
atmosphere at a pressure of 30 bar and temperatures from 25 to 300 °C. Structural analysis
using powder X-ray diffraction (PXRD) data suggested that hydrogenations in a temperature
range 25 – 200 ºC afford a uniform hydride phase with an orthorhombic structure (Immm, a ≈
4
.40 Å, b ≈ 3.97 Å, c ≈ 19.8 Å), whereas at 300 ºC mixtures of two orthorhombic phases with
c ≈ 19.86 and ≈19.58 Å were obtained. The assignment of a composition Eu Si H is based
3
4
2+x
on first principles DFT calculations, which indicated a distinct crystallographic site for H in
the Eu Si structure. In this position, H atoms are coordinated in a tetrahedral fashion by Eu
3
4
atoms. The resulting hydride Eu Si H is stable by -0.46 eV/H atom with respect to Eu Si
3
4
2
3
4
and gaseous H . Deviations between the lattice parameters of the DFT optimized Eu Si H
2
3
4
2
structure and the ones extracted from PXRD patterns pointed to the presence of additional H
in interstitials also involving Si atoms. Subsequent DFT modeling of compositions Eu Si H
3
3
4
and Eu Si H showed considerably better agreement to the experimental unit cell volumes. It
3
4
4
was then concluded that the hydrides of Eu Si have a composition Eu Si H (x < 2) and are
3
4
3
4
2+x
disordered with respect to H in Si Eu interstitials. Eu Si is a ferromagnet with a T at about
2
3
3
4
C
1
20 K. Ferromagnetism is effectively quenched in Eu Si H . The effective magnetic
3 4 2+x
2
+
7
moment for both materials is 7.5 µ which is typical for compounds containing Eu 4f ions.
B
Keywords: Zintl phases, Zintl phase hydrides, rare earth metal silicides, magnetic properties
2

ACCEPTED MANUSCRIPT
1
. Introduction
Zintl phases, which are composed of an active metal (i.e. alkali, alkaline earth, or rare earth)
and a more electronegative p-block metal or semimetal, represent a large family of inorganic
compounds [1-3]. As their characteristic feature, atoms of the electronegative component
appear reduced and may form polyanionic structures to achieve an octet. Like many
intermetallic compounds Zintl phases can react with hydrogen to form hydrides. However, the
rather high ionicity of Zintl phases makes such hydrides peculiar. Hydrogen takes an
ambivalent role and can be incorporated in two principal ways: either hydridic, where H is
exclusively coordinated by active metals (interstitial hydrides), or as part of the polyanion
where it acts as a covalently bonded ligand (polyanionic hydrides) [4,5]. Chemical structures
and physical properties of Zintl phases can change profoundly upon H incorporation. This
provides interesting prospects for fundamental inorganic chemistry and materials science.
Well investigated are the hydrogenation induced metal–semiconductor transitions for the
systems MTrTt – MTrTtH (M = alkaline earth; Tr = Al, Ga, In; Tt = Si, Ge, Sn). Here charge
imbalanced AlB -type related Zintl phases form charge balanced semiconductor hydrides in
2
which H is bonded to Tr and thus part of the polyanion [6,7]. Further, it has been shown that
CrB-type related Zintl phases MTt can be hydrogenated to yield hydrides MTtH with x
1
−
+x
3
2−
~
0.33, ~0.67, and ~0.87 featuring novel ribbon- and chain-like polyanions [Tt H] , [Tt H] ,
3 2
−
and [TtH] in which H is covalently bonded to a Tt element [8,9]. The discovery of
polyanions with Tt–H bonds represented an important extension of Zintl phase hydrides.
Another peculiar feature of MTtH1+x is the simultaneous presence of interstitial H which is
exclusively coordinated by M and not bonded to Tt. The CrB structure type is also adopted for
most rare earth (RE) monogallides, REGa, which – considering RE as trivalent – are
isoelectronic to MTt. In contrast with MTt, the REGa compounds display cooperative
magnetism. Recent studies on NdGa and GdGa showed that, similar to MTt, also REGa
transform into hydrides REGaH1+x by incorporating H both interstitially (i.e. exclusively
coordinated by RE atoms) and as part of a two-dimensional Ga-H polyanion. At the same time
the magnetic interaction is changed from ferromagnetic to antiferromagnetic nature
[
10,11,12].
In this paper we report on the hydrogenation behavior of Eu Si . Eu Si was first described in
3
4
3
4
2
004 [13]. Its body centered orthorhombic crystal structure (space group Immm, #71) is
depicted in Figure 1 [14]. Si atoms form hexagon rings, which in turn are condensed into one-
dimensional ribbons that run in the b direction (Figure 1a). Following the Zintl concept, three-
bonded (3b) Si atoms (henceforth termed Si1) carry a formal charge of -1, and two-bonded
(2b) ones (henceforth termed Si2) carry a charge of -2. A polyanionic strand is composed of
four Si atoms, 2 Si1 and 2 Si2 (there are two equivalent strands in the body centered unit cell).
Thus, a strand [Si1 Si2 ] carries a formal charge of -6, which is balanced by Eu cations,
2
2
provided they are in a +2 state. The divalent character of Eu was confirmed from magnetic
7
measurements, which showed a magnetic moment in agreement with a 4f electron
configuration when Eu Si is in the paramagnetic state above 117 K [13]. Eu1 atoms are
3
4
situated between Si hexagon rings. Eu2 atoms form arrays of edge condensed tetrahedra,
which separate Si hexagon ribbons in the c direction. Both Eu atoms together provide a
3

ACCEPTED MANUSCRIPT
trigonal prismatic coordination to Si atoms (Figure 1b). There is an apparent close relationship
to the orthorhombic CrB structure (space group Cmcm #63), which is adopted by the REGa
compounds (Figure 1b). Their structure is built from slabs of trigonal prims formed by the RE
atoms, which host Ga atoms that are arranged as zigzag chains. The Eu Si structure can be
3
4
considered as built up from slabs of CrB structure (with a thickness b/2) with alternating [010]
and [0-10] orientation. Condensing slabs at layers of common RE atoms will then connect
zigzag chains into hexagon rings.
Eu Si displays two magnetic transitions at 117 and 40 K [13]. The first one is attributed to a
3
4
ferromagnetic ordering of the Eu2 atoms. This transition probably relates to the ferromagnetic
transition in REGa compounds because RE atoms in REGa have the same structural
arrangement as the Eu2 atoms in Eu Si . The Curie temperatures T of REGa are in a range
3
4
C
1
5 – 190 K [15,16,17]. The second transition in Eu Si is caused by a ferromagnetic ordering
3 4
of the Eu1 atoms, resulting in a net ferrimagnetic ground state. The ferromagnetic properties
of Eu Si must be a consequence of the coupling of magnetic moments arising from localized
3
4
4
f electrons, which is mediated via conduction electrons. The validity of such a RKKY
coupling mechanism, however, contradicts the description of Eu Si as a charge balanced
3
4
Zintl phase, because in this simple picture all valence electrons are localized as bonds and
lone pairs within the Si polyanion, which would not leave any excess electrons for mediating
ferromagnetic coupling. Again, this is similar to ferromagnetic REGa for which H
incorporation triggers a transition from ferromagnetic to antiferrmomagnetic behavior [10,11].
In contrast with Eu Si , all RE atoms in REGa are crystallographically equivalent.
3
4
4

ACCEPTED MANUSCRIPT
2
. Methods
2
.1. Synthesis
All steps of synthesis and sample preparation were carried out in an Ar-filled glove box. Eu
99.9%) and Si (99.999%) were purchased from ABCR. Eu Si was prepared by arc-melting
(
3
4
stoichiometric mixtures of Eu and Si. To ensure homogeneity, samples were re-melted five
times and flipped between each melting. Slight changes in the stoichiometry or insufficient re-
melting led to the presence of EuSi and EuSi in the product. Most homogeneous samples
2
Eu Si were obtained when pre-reacting stoichiometric amounts of Eu and Si - enclosed in a
3
4
Ta ampule - in an RF furnace for 2 h at about 1000 °C into a mixture of Eu Si , EuSi, and
3
4
EuSi , and subsequently arc-melting this mixture. Buttons of Eu Si from arc-melting were
2
3
4
ground in an agate mortar and portions of 50–100 mg were pressed into pellets. The pellets
were placed in an Al O (corundum) crucible inside a custom-made stainless steel autoclave,
2
3
which was subsequently pressurized to 30 bar with H gas. The autoclave was then placed
2
inside a tube furnace and hydrogenations performed at room temperature, 100, 200, and 300
°
C for 24 h. The temperature inside the autoclave was monitored by a K-type thermocouple.
Both Eu Si and products obtained from its hydrogenation have a grey color and decompose
3
4
when exposed to humid atmosphere. For purity check and phase analysis of samples, powder
X-Ray diffraction (PXRD) was employed.
2
.2. Pressure Composition Isotherm
Hydrogen absorption properties were measured by Pressure-Composition-Isotherm (PCI)
curves with a commercial volumetric device (SETARAM PCT PRO) equipped with
calibrated and thermalized volumes and pressure gauges at 25 °C. The sample holder
corresponded to a stainless steel container closed with a metal seal and was inserted in a
resistance furnace at 250 and 300 °C. High purity hydrogen (6N) was introduced step by step
up to around 3 bar. Due to very slow kinetics, the time allowed for equilibrium was between
1
500 and 2000 minutes per pressure point. It was compulsory to load the sample in an Ar-
filled glove box in order to avoid air-exposure since surface oxidation resulted in a dramatic
decrease of kinetics, with sometimes no absorption.
2
.3. Structural Characterization of Eu Si and hydrogenous Eu Si
3 4 3 4
PXRD patterns were collected on a Panalytical X’Pert PRO diffractometer operated with Cu
either Cu Kα (λ = 1.5406 Å) radiation in θ-2θ diffraction geometry. Powder samples were
1
mounted on a Si wafer zero-background holder and sealed between kapton tape to ensure an
oxygen/moisture free atmosphere during data collection. Data were measured in a 2θ range 10
–
90°. High-energy X-ray diffraction experiments were carried out for Eu Si and products
3 4
from hydrogenations at 200 and 300 ºC (Eu3Si4_H2_200C and Eu3Si4_H2_300C) at the
beamline P02.1 at PETRAIII, DESY, which operates at a fixed energy of approximately 60
keV. The wavelength was determined to be 0.20727(6) Å by using a LaB NIST standard.
6
Powder samples were loaded in a glass capillary with 0.5 mm diameter. 2D diffraction
images, each obtained through the accumulation of 20 frames with an exposure time of one
second per frame, were collected with a Perkin Elmer amorphous silicon area detector
5

ACCEPTED MANUSCRIPT
(XRD1621) placed at 329 mm from the sample. The 2D diffraction images were then
integrated into a linear scattering signal with the software Fit2D and averaged [18]. The
Rietveld method [19] as implemented in the FULLPROF program (v. 2.05, July 2011) [20]
was used for structure and phase analysis. Structure refinements of synchrotron data were
based on the established structure model for Eu Si . A six-coefficient polynomial refinement
3
4
was used for the background. Lattice parameters were fitted with the value from structure
model. The peak shape was modeled by Thompson-Cox-Hastings pseudo-Voigt axial
divergence asymmetry. Fits could be considerably improved when applying a preferred
orientation correction to the intensities parallel to (101).
2
.4. In-situ diffraction
In-situ Synchrotron PXRD experiments were performed at the I11 beamline at the Diamond
light source, UK using a wavelength of 0.49404 Å (as determined by a Si standard) and a
wide-angle position sensitive detector based on Mythen-2 strip modules. The sample was
loaded into a 0.7 mm quartz capillary that was mounted on a Norby type cell [21] and sealed
using Epoxy. The cell was then mounted and leak checked with He gas before applying a
pressure of 30 bar H . Data was collected at 20 s per scan during a heat-cool temperature
2
program (RT – 400 °C – RT at 5 °C/min) using a hot-air blower. Consequently, the duration
of the heating and cooling ramps were about 80 min. In between the ramps a short period of
dwelling (5 min) was inserted. Data analysis was done by the Rietveld method implemented
in the software TOPAS 6 academic [22] in sequential mode. Peak functions were described by
th
the fundamental parameters approach and the background by a 6 degree Chebychev
polynomial. During sequential mode, unit cell parameters of three phases were refined (initial
Eu Si , hydrogenous Eu Si , and the impurity phase EuSi ) along with scale factors and
3
4
3
4
2
background parameters. All other parameters were kept fixed.
2
.5. Computations
DFT calculations were performed using the Vienna ab initio Simulation Package (VASP)
23,24] utilizing the projector augmented wave method (PAW) [25,26] to treat the interactions
[
between the electrons and the nuclei. The generalized gradient approximation (GGA) in the
parametrization of the Perdew-Burke-Ernzerhof (PBE) [27] approach was employed to
approximate the exchange and correlation. The plane wave basis set was terminated at a
kinetic energy cutoff of 400 eV. The conjugate gradient algorithm was used to relax the
atomic nuclei positions to a local minimum in the total energy landscape until the global break
−
3
condition of maximum 10 eV/Å of force between the atoms was reached. A 25×25×7
Monkhorst-Pack k-point mesh was used to sample the Brillouin zone[28] . Formation energies
∆E, referring to zero kelvin were calculated according to
∆
E= (1/n) [E(Eu Si H ) – E(Eu Si ) – n/2 E(H )]
3
4
n
3
4
2
where E denotes the total energy of the enclosed-in-parentheses system and n = 2,3,4. For the
3
H molecule, a box of 8×8×8Å dimensions was used with Γ-point sampling of the Brillouin
2
zone. Structure relaxations were carried out for the compositions Eu Si H , Eu Si H and
3
4
2
3
4
3
6

ACCEPTED MANUSCRIPT
Eu Si H using an orthorhombic starting cell with Z = 2. No symmetry restrictions were
3
4
4
applied during relaxation and the Findsym program [30] was used to check for symmetry of
–
3
atom arrangements after relaxations. Forces were converged to better than 10 eV/Å. All
materials were treated as non-magnetic and calculations were thus non spin-polarized.
2
.6. Magnetic Measurements
Magnetization measurements were performed as a function of temperature using a Quantum
Design MPMS SQUID magnetometer. M vs. T measurements were done both at low field (H
=
4 kA/m) in order to determine the transition temperature of the samples, as well as in high
field (H = 80 kA/m) in order to do a Curie-Weiss fit and estimate the effective bohrmagneton
number (µ ) and Curie-Weiss temperature (Θ ) of the samples. M vs. H measurements
eff
CW
were performed at 6K in the field range of ±7200 kA/m using a Quantum Design PPMS VSM
magnetometer to determine the saturation magnetization as well as study the field dependence
of the sample.
7

ACCEPTED MANUSCRIPT
3
. Results and Discussion
.1. Hydrogenation behavior of Eu Si4
3
3
Initially Eu Si was exposed to a hydrogen atmosphere of 30 bar during 24 h at various
3
4
temperatures in order to explore its hydrogen uptake behavior. Figure 2 shows selected PXRD
patterns of reaction products. Pristine Eu Si is remarkably susceptible to H uptake. As a
3
4
matter of fact, kinetic barriers appear to be small because almost quantitative hydride
formation is already observed at room temperature. At 100 °C and 200 °C a phase pure
hydride was obtained. As will be explained later, we assign this hydride a composition
Eu Si H . After hydrogenation at 300 °C a changed PXRD pattern is observed. Notably,
3
4
2+x
several reflections seem to be split, which may indicate the formation of another hydride.
When hydrogenating at temperatures above 300 °C, the crystallinity of products appears
drastically reduced. Above 500 °C the product contained a substantial fraction EuSi2,
indicating H-induced oxidative decomposition. In this work we restrict ourselves to hydride
phases obtained at low hydrogenation temperatures.
The diffraction patterns of Eu Si H from 100 °C and 200 °C hydrogenations (samples
3
4
2+x
Eu Si _H _100C and Eu Si _H _200C, respectively) could be readily indexed in a body-
3
4
2
3
4
2
centered orthorhombic cell with lattice parameters a ≈ 4.39 Å, b ≈ 3.97 Å, c ≈ 19.8 Å. As a
matter of fact, these parameters resemble those of the starting material Eu Si (a = 4.6103(3)
3
4
Å, b = 3.9574(1) Å, c = 18.2239(4) Å) [13] which suggests a close structural relationship
between Eu Si H and Eu Si . In-house PXRD data was not suitable for Rietveld
3
4
2+x
3
4
refinement. PXRD data collected at a synchrotron were of vastly superior quality.
Synchrotron PXRD patterns of the starting material and of the hydrogenated samples at 200
°
C and 300 °C are shown in Figure 3. The pattern of the 300 °C product (sample
Eu Si _H _300C) could be satisfactorily refined as a mixture of two orthorhombic phases.
3
4
2
The first phase corresponds well to the hydride phase obtained in the 200 °C experiment. The
only difference is that the c parameter appears slightly increased, from 19.81 to 19.86 Å. The
second phase has a significantly increased a (~4.44 Å) and decreased c parameter (~19.58 Å).
In the following we designate this phase as Eu Si H . Rietveld plots are shown as supporting
3
4
~2
information, Figure A1. The refinement results are contained in Tables 1 and 2 which also
include the structure parameters of Eu Si obtained from single crystal X-ray diffraction data,
3
4
as reported in the literature [13]. There is good agreement between our parameters from
powder refinement and the reported ones. Table A1 (supporting information) lists interatomic
distances.
To elucidate the hydrogenation pathway and to possibly shed light into the formation of two
phases when hydrogenating at 300 °C, an in-situ synchrotron PXRD experiment was
performed. An overview of the results from this experiment can be seen in Figure 4 and
Table 3. In contrast with the starting material used for the autoclave hydrogenation
experiments, the Eu Si sample synthesized for the in-situ experiment contained 6.1% EuSi
2
3
4
impurity (see Figure A2 and Table A2, supporting information). EuSi does not absorb
2
hydrogen and its presence did not hamper the analysis of data. The sample was subjected to a
hydrogen atmosphere of 30 bar and subsequently heated to 400 °C at a rate of 5 °C/min,
8

ACCEPTED MANUSCRIPT
dwelled for 5 min, and then cooled back to room temperature at a rate of 5 °C/min. (we
remind that autoclave experiments applied a dwelling time of 24 h). A small but significant
increase in the unit cell volume was noticed already at the start of the experiment. This
indicates that Eu Si absorbs hydrogen at room temperature, which is in agreement with the
3
4
observation from the autoclave experiments (cf. Figure 2).
Upon heating, the actual transformation into a hydride phase started at 90 °C (after 0.25 h) –
as indicated by the appearance of additional diffraction lines – and was completed at 140 °C
(
after 0.42 h), as indicated by the vanished reflections from Eu Si . The lattice parameters of
3 4
the obtained hydride phase are a ≈ 4.4 Å, b ≈ 4 Å, c ≈ 19.9 Å and thus (considering also
thermal expansion) relate well to those of phase Eu Si H obtained from the autoclave
3
4
2+x
hydrogenations. Interestingly, upon heating beyond 180 °C (i. e. after 0.55 h) the c axis of the
hydride phase contracted continuously, and was reduced by 0.6 Å when reaching 400 °C. The
c axis contraction was paralleled by a slight expansion of the a axis (by 0.1 Å). This behavior
was reversible upon cooling. Figure 5 depicts some detailed diffraction patterns. At 100 °C
the pattern represents a phase mixture between Eu Si and Eu Si H . Characteristic of the
3
4
3
4
2+x
-1
latter phase is the pair of reflections centered at Q ≈ 2.2 Å where Eu Si has a single
3
4
reflection. The diffraction patterns at 250 °C upon heating and cooling appear virtually
identical. The pattern obtained at 400 °C yields the lattice parameters a ≈ 4.51, b ≈ 4.0, c ≈
1
9.31 Å, which are rather different from Eu Si H . Back at room temperature, values of the
3 4 2+x
lattice parameters corresponded closely to those of Eu Si H obtained in the autoclave
3
4
2+x
experiment at 300 °C (i.e. sample Eu Si _H _300C, cf. Tables 1 and 3).
3
4
2
To sum up the salient observations with the in-situ experiment: there is an initial formation of
a hydride in the temperature interval 90 – 140 °C, which upon further heating displays
continuous variation of the c and a lattice parameters. These variations are reversible and the
initially formed hydride is regained when cooling to room temperature. We conjecture that the
initially formed phase corresponds to Eu Si H obtained in autoclave hydrogenation
2+x
3
4
experiments at 100 – 200 °C. At higher temperatures the hydrogen content is reduced and the
phase seen at 400 °C in the in-situ experiment relates to second phase obtained in the 300 °C
autoclave hydrogenation experiment, Eu Si H . For unknown reasons, a quenchable form of
3
4
~2
Eu Si H can only be obtained after prolonged sintering at temperatures above 200 °C.
3
4
~2
As a next step we investigated the thermal stability/desorption behavior of Eu Si H by
3
4
2+x
exposing parts of the sample Eu Si _H _200C to a dynamic vacuum at elevated temperatures.
3
4
2
Results are compiled in Figure A3 in the supporting information. The Eu Si H phase was
3
4
2+x
essentially maintained up to 200 °C. At 300 °C a changed diffraction pattern was obtained
with a weaker crystallinity. Interestingly, there is strong indication that Eu Si H partially
3
4
~2
formed during the 300 °C desorption experiment. Eu Si did not reform in desorption
3
4
experiments. Thus, the hydrogenation of Eu Si to Eu Si H is not reversible. Raman and IR
3
4
3
4
2+x
spectra of Eu Si and Eu Si H are essentially featureless which indicates that both phases
3
4
3
4
2+x
are metals.
Pressure composition isotherms (PCIs) measurements were attempted in order to determine
the hydrogen content of Eu Si H . It was found that the sample absorbs hydrogen within a
3
4
2+x
9

ACCEPTED MANUSCRIPT
single step (single plateau pressure) below 3 bar at 250 and 300 °C and the equilibrium
pressure of the absorption plateau is very low and therefore below the detectable limits of the
instrument. The maximum hydrogen capacity only reached 1.7 H/f.u. under these conditions,
most probably due to kinetic limitation (see Figure A4). Diffraction analysis of the PCI
samples conducted at 250 and 300 °C showed the presence of a phase mixture, analogous to
the sample Eu Si _H _300C (Table A3). This is not surprising since the equilibrium times
3
4
2
during PCI measurements (~2000 min) are similar to the dwelling time applied in the
autoclave hydrogenation experiments (24 h), thus supporting the conjecture that the formation
of a quenchable phase Eu Si H is associated with a slow kinetics.
3
4
~2
3
.2. Structural variations and energetics of H incorporation
In the following we compare the structures of Eu Si H and Eu Si (cf. Tables 1-3). The a
3
4
2+x
3
4
lattice parameter is the stacking direction of hexagon ring ribbons. With respect to Eu Si this
3
4
parameter is contracted by about 5% for Eu Si H (from 4.62 to 4.4 Å) and by about 4% for
3
4
2+x
Eu Si H . The b parameter (i.e. the direction of ribbons) is virtually unaffected upon hydride
3
4
~2
formation, whereas the c parameter increases considerably, from 18.24 to 19.81 – 19.87 Å for
Eu Si H and to 19.58 Å for Eu Si H . The associated volume increase of the unit cells are
3
4
2+x
3
4
~2
about 4.2 and 3.5%, respectively.
The high neutron capture cross section of Eu makes the application of neutron diffraction for
1
53
structural analysis of Eu Si H difficult, requiring samples with isotopically enriched Eu.
3
4
2+x
For identifying possible locations for H atoms, we therefore utilized our experience with
REGaH2-x [10,11]. The precursor Zintl phases REGa crystallize with the CrB structure
[
15,16,17]. As discussed in the introduction, the Eu Si structure can be considered built from
3 4
slabs of CrB structure. The arrangement of Eu2 atoms in “Si” free layers corresponds to an
array of edge sharing tetrahedra (cf. Figure 1). The centers of these tetrahedra are occupied in
REGaH2-x by one kind of H atoms (H1). This is similar to hydrides with the ZrCuSiAs
structure type (e.g. CeCoSiH) [31]. Assuming the same scenario for Eu Si one arrives at an
3
4
interstitial hydride Eu Si H . The centers of Eu2 tetrahedra corresponds to a site 4i (0,0,z) z ≈
3
4
2
4
0
.25 in space group Immm. The structure parameters of the Eu Si H model were
3 4 2
subsequently optimized by non spin polarized DFT calculations. To validate the simplified
non spin polarized approach, we also subjected the Eu Si structure to DFT optimization. The
3
4
results are given in Tables 4 and A4. We note that the Immm structure of Eu Si is excellently
3
4
reproduced by DFT structure optimization (cf. Tables 1, 2, 4).
The DFT optimized structure for the Eu Si H model (space group Immm) is shown in Figure
3
4
2
6
. The incorporation of H into the Eu2 tetrahedral site in Eu Si is clearly energetically
4 3 4
favorable. The calculated energy for the reaction Eu Si + H → Eu Si H is -0.918 eV.
3
4
2
3
4
2
However, the stabilization E/H = -0.459 eV is lower compared to REGaH2-x (E/H1 = -0.76
eV/H) because in the latter, H (in the equivalent H1 position) becomes situated in an
environment of trivalent RE. Further, the DFT optimized structure of Eu Si H structure
3
4
2
shows evidently the experimentally observed trend in the lattice parameter variations, that is, a
contracted a, constant b and expanded c parameter with respect to Eu Si . Importantly, the
3
4
values are rather close to those of Eu Si H from which we infer the approximate
3
4
~2
1
0

ACCEPTED MANUSCRIPT
composition of this phase. At the same time we are unable to provide an explanation as to
why a quenchable form of Eu Si H was only obtained in sintering experiments.
3
4
~2
In contrast with the other RE metals, Eu can easily change between the di- and trivalent state.
Several scenarios are possible when going from Eu Si to Eu Si H . Importantly, in interstitial
3
4
3
4
2
hydrides H behaves typically hydridic and, thus, interstitial hydride formation can be
2
+
accompanied by both, oxidation of the polyanion and oxidation of Eu . This is sketched in
2
+
Figure 7. If Eu was oxidized, a block [Eu2 H ] would carry four positive charges and no
2
2
+
2
changes are expected for the polyanion. If Eu present in Eu Si is not oxidized, a block
3
4
[
Eu2 H ] carries two positive charges. Assuming charge balance, the Si hexagon chain
2 2
becomes oxidized and may respond to this by developing π−bonding. This scenario is
indicated when comparing the Si-Si distances in the DFT optimized structures of Eu Si and
3
4
Eu Si H (Table A4). Both distances, Si1-Si2 and Si2-Si2 reduce slightly, by 0.02 – 0.03 Å,
4
4
2
when going from Eu Si to Eu Si H .
3
4
3
4
2
The considerably larger c parameter of Eu Si H compared to Eu Si H (and compared to
3
4
2+x
3
4
~2
the DFT optimized structure for Eu Si H ) point to the incorporation of additional H in
3
4
2
◦
Eu Si , that is, a composition Eu Si H is attained upon hydride formation below 200 C.
3
4
3
4
2+x
This situation would remind of REGa for which additional H, on the position H2, is inserted
between Ga atoms of neighboring zigzag chains [10,11]. The H2 site is only partially
occupied (2/3) in the hydrides REGaH1.66. Importantly, H1 (fully occupied) and H2 are
chemically different. The former is interstitial, primarily leading to an oxidation of the
polyanion, the latter is a part of the polyanion [10].
Thus, again guided by REGa, we indentified the 4i position (0,0,~0.87) between two Si2
atoms of neighboring hexagon rings as a possible location for additional H. This position
centers at the same time Eu2 Eu1 triangles (cf. Figure 1a). The resulting Si Eu trigonal
2
2
3
bipyramidal environment is comparable to the Ga La environment in the recently reported H-
2
3
disordered hydride LaGa D , which is based on the AlB structure type [32]. Subsequent
2
0.7
2
DFT optimization of the structure parameters yielded a monoclinic structure for the hydride
Eu Si H . The relaxed structure of this hydride is shown in Figure 8a. It features a polyanion
3
4
4
4
-
[
Si1 Si2 H2 ] in which H2 is regularly bonded to Si2 (d(Si2-H2) = 1.61 Å). At the same
2 2 2
time, planar hexagon rings corrugate slightly and attain a chair conformation. The situation is
strikingly similar to recently discovered BaSiH and SrSiH5/3-x with polyanionic zigzag
2
-x
−
2−
chains [SiH] and ribbons [Si H] , respectively [8,9]. In these compounds (calculated) Si-H
2
distances are in a range 1.64 to 1.66 Å.
The reaction energy for the step Eu Si H + H → Eu Si H is still negative, -0.395 eV, but it
3
4
2
2
3
4
4
is clear that the H2 position is far less favorable compared to H1. The structure of Eu Si H is
3
4
4
monoclinic, but the lattice metrics is close to orthorhombic. The monoclinic angle of a body
centered lattice, which is preferably used when comparing to the Immm structures of Eu Si4
3
and Eu Si H , is 91.9° (cf. Table 4). The c lattice parameter of Eu Si H is considerably
3
4
2
3
4
4
increased compared to Eu Si H , to 19.88 Å, whereas the a and b parameters are essentially
3
4
2
the same. Thus, it appears that a closer agreement to the experimental values was obtained (cf.
Table 1). We also considered an ordered model Eu Si H with half of the Si2 atoms
3
4
3
1
1

ACCEPTED MANUSCRIPT
terminated by H2 type atoms. Its DFT optimized structure, which also is monoclinic, is shown
in Figure 8b. The increase of the c parameter with respect to Eu Si is less pronounced (19.61
3
4
Å).
The monoclinic symmetry of the calculated models Eu Si H and Eu Si H is a consequence
3
4
4
3
4
3
of the puckering of hexagon layers. The symmetry reduction is clearly not observed in the
PXRD pattern of e.g. Eu Si H _H _200C (cf. Figures 3 and A1). Therefore it can be
3
4
2
2
concluded that the additional H in Eu Si H probably incorporates in a more complicated,
3
4
2+x
disordered, fashion. A similar situation has been described for LaGa D [32].
2
0.7
3
.3. H-induced magnetic property changes
The ZFC and FC magnetization of Eu Si and Eu Si H (sample Eu Si _H _200C) as a
3
4
3
4
2+x
3
4
2
function of temperature in an applied field of 4 kA/m is shown in Figure 9a. The Zintl phase
Eu Si has two noteworthy features, a sharp increase in the magnetization at 120 K and
3
4
another pronounced feature at about 40 K. In Ref. [13] the same behavior was observed, and
the authors suggested a ferromagnetic ordering at 117 K for the Eu2 site and a consecutive
ferromagnetic ordering of the Eu1 site at about 40 K. However, with a weak antiferromagnetic
coupling between the two substructures which gives rise to a ferrimagnetic total structure at
low temperatures. Upon hydrogenation it can be observed that the ferromagnetic transition at
1
20 K disappears, indicating that the hydrogenation disrupts the ordering of the Eu2 moments
at 120 K. Since SQUID magnetometry is very sensitive to magnetic impurities (especially
ferromagnetic ones) it is also clear that there is no trace of the original ferromagnetic Eu Si4
3
phase in the hydride sample.
The ZFC and FC thermomagnetic curves for the Eu Si H sample in Figure 9a indicates a
3
4
2+x
para- to antiferromagnetic transition at about 40 K. However, with some uncompensated spins
in the antiferromagnetic phase giving rise to an excess moment as seen from the separation of
the ZFC and FC curves below the transition temperature. The magnetization as a function of
field behavior at 6 K is complex and exhibits a second magnetic hysteresis loop at high fields
(above 3000 kA/m), see Figure 9b). A similar behavior to the data presented in Figure 9b has
been observed in single crystals of the antiferromagnet UIrSi , which crystallizes in the
3
tetragonal BaNiSn3 structure type [33]. The magnetic moments in UIrSi3 order
ferromagnetically in the ab plane with the moments aligned along the c axis. The
ferromagnetic planes order then antiferromagnetically with respect to each other, yielding a
net antiferromagnetic structure. Upon application of a sufficiently large magnetic field along
the c axis (easy axis), UIrSi exhibits a metamagnetic transition to a ferromagnetic
3
configuration. No metamagnetic transition was observed upon application of the magnetic
field along the a axis. Even though the behavior observed in UIrSI agrees well with the data
3
presented in Figure 9b (if the polycrystallinity of the Eu Si H sample is taken into account),
3
4
2+x
it cannot with certainty be concluded that this is the origin of the observed effect from the
collected data. From Curie-Weiss fits it was found that p ≈ 7.5 µ (~7.2 for Eu Si and ~7.7
eff
B
3
4
7
2+
2+
3+
for the hydrogenated sample), indicating 4f Eu for both samples (Eu ~8 and Eu ~3.4
µ ). The Curie-Weiss temperatures (Ω ) are ~110 and ~5 K for Eu Si and Eu Si H
2+x
B
CW
3
4
3
4
1
2

ACCEPTED MANUSCRIPT
respectively. The drastic change in ΩCW further underlines the disruptive effect that hydrogen
has on the magnetic interaction in Eu Si .
3
4
4
. Conclusions
Eu Si absorbs hydrogen already at room temperature when exposing the material to a
3
4
hydrogen atmosphere of 30 bar. Structural analysis by a combination of X-Ray powder
diffraction and DFT modeling suggests that at temperatures up to 200 °C a hydride Eu Si H
2+x
3
4
is formed, whereas sintering at higher temperatures (300 °C) affords a hydride with lower
hydrogen content, Eu Si H . Eu Si H desorbs H irreversibly above T = 200 °C. The
3
4
~2
3
4
2+x
ferromagnetic properties of Eu Si are effectively quenched upon hydrogenation to
3
4
Eu Si H . The effective bohrmagneton number of Eu Si H is close to that of Eu Si above
3
4
2+x
3
4
2+x
3
4
Tc, indicating that Eu possesses the same oxidation state (i.e. +2) in both systems.
Acknowledgement. This work has been supported by the Swedish research council and the
NordForsk project “Neutrons for multifunctional hydrides (FunHy)”.
1
3

ACCEPTED MANUSCRIPT
References
[
1
1] S. M. Kauzlarich, Chemistry, Structure, and Bonding of Zintl Phases and Ions, Wiley,
996.
[
(
2] R. Nesper, The Zintl-Klemm Concept - A Historical Survey, Z. Anorg. Allg. Chem. 640
2014) 2639–2648. doi:10.1002/zaac.201400403.
[
3] S. C. Sevov, Zintl phases, in: J. H. Westbrook, R. L. Fleischer (Eds.), Intermetallic
Compounds: Principles and Practice, John Wiley and Sons Ltd, 2002, pp. 113–132.
[
4] U. Häussermann, V. F. Kranak, K. Puhakainen, Hydrogenous Zintl Phases: Interstitial
Versus Polyanionic Hydrides, in: T. F. Fässler (Ed.), Zintl phases: Principles and recent
developments, Structure and Bonding 139 (2011), 143–161. doi: 10.1007/430_2010_20.
[
5] U. Häussermann, Coexistence of hydrogen and polyanions in multinary main group
element hydrides, Z. Kristallogr. 223 (2008) 628–635. doi:10.1524/zkri.2008.1016.
[
6] T. Björling, D. Noreus, K. Jansson, M. Andersson, E. Leonova, M. Eden, U. Hålenius,
U. Häussermann, SrAlSiH: A Polyanionic Semiconductor Hydride, Angew. Chem. Int. Ed.
4 (2005) 7269–7273. doi:10.1002/anie.200502090.
4
[
7] M. H. Lee, T. Björling, B. C. Hauback, T. Utsumi, D. Moser, D. Bull, D.
Noreus, O. F. Sankey, U. Häussermann, Crystal structure, electronic structure, and
vibrational properties of MAlSiH (M=Ca,Sr,Ba): Hydrogenation-induced semiconductors
from the AlB2-type alloys MAlSi, Phys. Rev.
doi:10.1103/PhysRevB.78.195209.
B
78 (2008) 195209.
[
8] H. Auer, R. Guehne, M. Bertmer, S. Weber, P. Wenderoth, T. C. Hansen, J. Haase,
H. Kohlmann, Hydrides of Alkaline Earth–Tetrel (AeTt) Zintl Phases: Covalent Tt–H Bonds
from Silicon to Tin, Inorg. Chem. 56 (2017) 1061–1071. doi:10.1021/acs.inorgchem.6b01944.
[
9] H. Auer, R. Schlegel, O. Oeckler, H. Kohlmann, Structural and Electronic Flexibility in
Hydrides of Zintl Phases with Tetrel – Hydrogen and Tetrel – Tetrel Bonds, Angew. Chem.
Int. Ed. 56 (2017) 12344–12347. doi:10.1002/anie.201706523.
[
10] J. Ångström, R. Johansson, T. Sarkar, M. H. Sørby, C. Zlotea, M. S. Andersson, P.
Nordblad, R. H. Scheicher, U. Häussermann, M. Sahlberg, Hydrogenation-Induced Structure
and Property Changes in the Rare-Earth Metal Gallide NdGa: Evolution of a [GaH]2-
Polyanion Containing Peierls-like Ga-H Chains, Inorg. Chem. 55 (2016) 345–352.
doi:10.1021/acs.inorgchem.5b02485.
[
11] R. Nedumkandathil, V. F. Kranak, R. Johansson, J. Angström, O. Balmes, M. S.
Andersson, P. Nordblad, R. H. Scheicher, M. Sahlberg, U. Häussermann, Hydrogenation
induced structure and property changes in GdGa, J. Solid. State Chem. 239 (2016) 184–191.
doi:10.1016/j.jssc.2016.04.028.
[
12] H. Auer, R. Nedumkandathil, U. Häussermann, H. Kohlmann, The Hydrogenation of the
Zintl Phase NdGa Studied by in situ neutron diffraction, Z. Anorg. Allg. Chem. 645 (2019)
75–181. doi:10.1002/zaac.201800459.
1
1
4

ACCEPTED MANUSCRIPT
[
13] F. Weitzer, Y. Prots, W. Schnelle, K. Hiebl, Y. Grin, The novel silicide Eu3Si4:
Structure, chemical bonding, magnetic behavior and electrical resistivity, J. Solid State Chem.
77 (2004) 2115–2121. doi:10.1016/j.jssc.2004.02.013.
1
[
14] The structure refinement reported in ref. [13] described one of the Si atoms (2b) Si2 with
a split position, indicating a slight corrugation of the Si hexagon rings. In the discussion of the
experimental Eu Si structure in this work an idealized structure with planar hexagon rings is
3
4
assumed. In this structure Si2 is located on a position 4i (0,0.4360).
[
15] B. Barbara, Nguyen-Van-Nhung, E. Siaud, Magnetic properties of rare earth-gallium
equiatomic compounds, CR. Acad. Sci B 274 (1972) 1053–1056.
[
16] H. Fujii, N. Shohata, T. Okamoto, E. Tatsumoto, Magnetic Properties of Rare Earth
Gallium Compounds RGa, J. Phys. Soc. Jpn. 31 (1971) 1592–1592.
doi:10.1143/JPSJ.31.1592.
[
17] N. Shohata, Magnetic Properties of Rare Earth Gallium Intermetallic Compounds, J.
Phys. Soc. Jpn. 42 (1977) 1873–1880. doi:10.1143/JPSJ.42.1873.
[
18] A. P. Hammersley, S. O. Svensson, M. Hanfland, A. N. Fitch, D. Hausermann, Two-
dimensional detector software: From real detector to idealised image or two-theta scan, High
Pressure Res. 14 (1996) 235–248. doi:10.1080/08957959608201408.
[
19] H. M. Rietveld, A profile refinement method for nuclear and magnetic structures, J.
Appl. Crystallogr. 2 (1969) 65–71. doi:10.1107/S0021889869006558.
[
20] R.-C. Juan, Fullprof: A Program for Rietveld Refinement and Pattern Matching Analysis,
Abstract of the Satellite Meeting on Powder Diffrac- tion of the XV Congress of the IUCr,
Toulouse, France 127 (1990).
[
21] P. Norby, Hydrothermal Conversion of Zeolites: An in Situ Synchrotron X-ray Powder
Diffraction Study, J. Am. Chem. Soc. 119 (1997) 5215–5221. doi:10.1021/ja964245g.
[
22] A. A. Coelho, Computer programs TOPAS and TOPAS-Academic: an optimization
program integrating computer algebra and crystallographic objects written in C ++ 210–218
(2018). doi:10.1107/S1600576718000183.
[
23] G. Kresse, J. Hafner, Ab initio molecular dynamics for liquid metals, Phys.
Rev. B 47 (1993) 558–561. doi:10.1103/PhysRevB.47.558.
[
24] G. Kresse, J. Furthmuller, Efficiency of ab-initio total energy calculations for
metals and semiconductors using a plane-wave basis set, Comput. Mater. Sci. 6 (1996) 15–50.
doi:10.1016/0927- 0256(96)00008-0.
[
1
25] P. E. Blöchl, Projector augmented-wave method, Phys. Rev. B 50 (1994) 17953–
7979. doi:10.1103/PhysRevB.50.17953.
[
26] G. Kresse, D. Joubert, From ultrasoft pseudopotentials to the projector augmented-wave
method, Phys. Rev. B 59 (1999) 1758–1775. doi:10.1103/PhysRevB.59.1758.
1
5

ACCEPTED MANUSCRIPT
[
27] J. P. Perdew, K. Burke, M. Ernzerhof, Generalized Gradient Approximation Made
Simple, Phys. Rev. Lett. 77 (1996) 3865–3868. doi:10.1103/PhysRevLett.77.3865.
[
28] J. P. Perdew, K. Burke, M. Ernzerhof, Generalized Gradient Approximation Made
Simple [erratum], Phys. Rev. Lett. 78 (1997) 1396–1396. doi:10.1103/PhysRevLett.78.1396.
[
1
29] H. J. Monkhorst, J. D. Pack, Special points for Brillouin-zone integrations, Phys. Rev. B
3 (1976) 5188–5192. doi:10.1103/PhysRevB.13.5188.
[
30] H. T. Stokes, D. M. Hatch, FINDSYM: program for identifying the space-group
symmetry of crystal, J. Appl. Crystallogr. 38 (2005) 237–238.
doi:10.1107/S0021889804031528.
a
[
(
32] R. Pöttgen, D. Johrendt, Materials with ZrCuSiAs-type Structure, Z. Naturforsch. 63b
2008) 1135–1148. doi:10.1515/znb- 2008-1001.
[
31] A. Werwein, C. Benndorf, M. Bertmer, A. Franz, O. Oeckler, H. Kohlmann,
Hydrogenation properties of LnAl (Ln = La, Eu, Yb), LaGa , LaSi and the crystal structure
2
2
2
of LaGa H0.71(2), Crystals 9 (2019) 1–12. doi:10.3390/cryst9040193.
2
[
33] J. Valenta, F. Honda, M. Valiska, P. Opletal, J. Kastil, M. Misek, M. Divis, L.
Sandratskii, J. Prchal, V. Sechovsky, Antiferromagnetism and phase transitions in
noncentrosymmetric UIrSi3, Phys. Rev. 97 (2018) 144423.
doi.org/10.1103/PhysRevB.97.144423
B
1
6

ACCEPTED MANUSCRIPT
Figures
Figure 1. a) Idealized crystal structure of Eu Si (space group Immm) with polyanionic
3
4
ribbons of planar hexagon rings projected approximately along the a (left) and the b (right)
direction [13,14]. The arrangement of Eu2 atoms into arrays of edge condensed tetrahedra and
the trigonal prismatic coordination of Eu2 around Si atoms are emphasized. b) Relation of the
CrB and Eu Si structures. Eu and Si atoms are depicted as grey and red circles, respectively.
3
4
Figure 2. PXRD patterns (Cu Kα ) of products from the hydrogenation of Eu Si at p = 30
1
3
4
bar during 24 h experiments at room temperature, 100 °C, 200 °C, and 300 °C. The pattern of
the Eu Si starting material is shown as black line.
3
4
Figure 3. Synchrotron PXRD patterns (λ = 0.2073 Å) for Eu Si and products obtained from
3
4
2
4 h hydrogenations at 200 °C (Eu Si H 200C) and 300 °C (Eu Si H 300C).
3 4_ 2_ 3 4_ 2_
Figure 4. a) Densiometric view of diffraction patterns during in-situ synchrotron measure-
ments. b) Unit cell parameters of the hydride phase obtained from sequential Rietveld
refinements.
Figure 5. Selected diffraction patterns during different stages of hydrogenation (as shown in
Figure 4). Reflections from the EuSi impurity phase are marked with asterisks (*).
2
Figure 6. DFT optimized crystal structure model for Eu Si H .
3
4
2
Figure 7. Different scenarios for products from Eu Si hydrogenation.
3
4
Figure 8. DFT optimized crystal structure models for Eu Si H , Eu Si H .
3
4
4
3
4
3
Figure 9. a) Zero field cooled and field cooled curves for Eu Si and Eu Si H in an applied
3
4
3
4
2+x
magnetic field H = 4 kA/m. b) Magnetization as a function of applied magnetic field for
Eu Si and Eu Si H at 6 K.
3
4
3
4
2+x
1
7

ACCEPTED MANUSCRIPT
Tables:
Table 1. Crystallographic information for Eu Si and hydrogenated Eu Si samples from
3
4
3
4
Rietveld refinement.
Table 2. Atomic coordinates and isotropic displacement parameters for Eu Si and
3
4
hydrogenated Eu Si samples.
3
4
Table 3. Cell parameters obtained by sequential Rietveld refinements of in-situ data.
Table 4. Parameters of DFT optimized structures.
1
8

ACCEPTED MANUSCRIPT
Ferromagnetic Eu
ferromagnetism.
3
Si
4
3 4
absorbs hydrogen to yield hydrides Eu Si H2+x with quenched
1

ACCEPTED MANUSCRIPT
Table 1: Crystallographic information for Eu
Rietveld refinements.
3
Si
4
and hydrogenated Eu
3
Si
4
samples from
3
2
χ
SG: Immm
a (Å)
b (Å)
c (Å)
V (Å )
Rbragg
Rf
Eu Si
4.6164(4)
3.9583(3)
18.229(1)
333.10(7)
3
4
(ref. [13])
Eu Si
4.61499(9) 3.95949(8) 18.2361(5) 333.23(1)
4.49
4.8
3.71 0.410
3.68 0.831
3
4
Eu Si _H _200C 4.40055(5) 3.97166(5) 19.8098(3) 346.225(8)
3
4
2
Eu Si _H _300C
0.941
4.06
3
4
2
Eu
3
Si
4
H
2+x
4.40247(7) 3.97478(9) 19.8646(5) 347.61(1)
4.4391(1) 3.9778(1) 19.5784(8) 345.72(2)
5.35
7.22
6
4(1) %
Eu Si
5(7)%
3
4
H
~2
6.01
3

ACCEPTED MANUSCRIPT
Table 2: Atomic coordinates and isotropic displacement parameters for Eu
3
Si
4
and hydrogenated
3 4
Eu Si samples.
2
Atom
Wyck
x
y
z
B(Å )
Eu1
Eu2
Si1
2a
4j
4i
4j
0
½
0
0
0
0
0
0
0.51(3)
0.54(2)
0.6(1)
Eu Si
ref. [13])
0.31500(6)
0.4360(4)
0.1377(3)
3
4
(
Si2
½
0.7(1)
Eu1
Eu2
Si1
2a
4j
4i
4j
0
½
0
0
0
0
0
0
1.03(2)
1.03(2)
1.69(7)
1.69(7)
Eu Si
0.31487(0)
0.43667(4)
0.13800(4)
3
4
Si2
½
Eu Si _H _200C
Eu1
Eu2
Si1
2a
4j
4i
4j
0
½
0
0
0
0
0
0
0.909(5)
0.909(5)
1.03(1)
1.03(1)
3
4
2
0.31019(0)
0.440164)
0.12066(4)
Si2
½
Eu
3
Si
Eu
4
_H
Si
2
_300C
Eu1
Eu2
Si1
2a
4j
4i
4j
0
½
0
0
0
0
0
0
0.601(7)
0.601(7)
0.23(2)
0.23(2)
3
4
H
2+x
0.31014(0)
0.44593(1)
0.12907(1)
Si2
½
Eu Si _H _300C
Eu1
Eu2
Si1
2a
4j
4i
4j
0
½
0
0
0
0
0
0
1.9(2)
1.9(2)
1.4(4)
1.4(4)
3
4
2
Eu Si H
0.31190(0)
0.43072(2)
0.11365(2)
3
4
~2
Si2
½

ACCEPTED MANUSCRIPT
Table 3: Cell parameters obtained by sequential Rietveld refinements of in-situ data.
3
Time (h) a (Å)
4.6146(4)
b (Å)
3.9572(3)
c (Å)
18.229(2)
V (Å )
Eu Si (93.9%)
332.90
3
4
Start
0
4.6193(1)
4.4371(3)
4.5085(2)
4.4397(2)
4.4013(2)
3.96156(9) 18.2454(5) 333.88
2
4
2
50 °C
00 °C
50 °C
0.78
1.33
1.87
2.63
3.9836(2)
3.9968(2)
3.9842(2)
3.9756(2)
19.785(1)
19.3128(8) 348.02
19.759(1)
19.857(1)
349.73
349.53
347.47
End

ACCEPTED MANUSCRIPT
Table 4: Parameters of DFT optimized structures.
Atom Wyck
3
x
y
z
a, b, c (Å) V(Å )
Eu1
Eu2
Si1
2a
4j
4i
4j
2a
4j
4i
4j
4i
2e
2e
2e
2e
2e
2e
2e
2e
2e
2e
2a
4i
0
½
0
½
0
½
0
½
0
0
0
0
0
0
0
0
0
0
¼
¼
¼
¼
¼
¼
¼
¼
¼
¼
0
0
4.5817
3.9445
18.2180
Eu Si4
Immm (71)
0.3156
0.4351
0.1383
0
0.3064
0.4399
0.1259
0.2567
0.2529
-0.0585
0.5560
0.8069
0.6876
0.1273
0.3768
-0.0045
0.5068
0.1286
0
0.6916
0.5593
0.8765
0.7450
0.1231
329.25
3
Si2
Eu1
Eu2
Si1
Si2
H1
Eu1
Eu21
Eu22
Si11
Si12
Si21
Si22
H11
H12
H2
4.4208
3.9716
19.3803
Eu Si H
2
3
4
Immm (71)
340.27
342.56
0.2396
0.7364
0.7483
0.2918
0.2444
0.7690
0.7492
0.2414
0.2458
0.1366
0
0.4836
0.0551
0.5204
-0.0070
0.1162
4.3903
3.9794
19.6102
Eu Si H
3
β = 90.9°
3
4
P2 /m (11)
1
Eu
3
Si
4
H
4
Eu1
Eu2
Si1
Si2
H1
4.4192
3.9625
19.8895
I2/m (12)
0
0
0
0
4i
4i
4i
4i
348.10
β = 91.9°
H2
0

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT